Hashing over Predicted Future Frames for Informed Exploration of Deep Reinforcement Learning.
Haiyan YinJianda ChenSinno Jialin PanPublished in: IJCAI (2018)
Keyphrases
- reinforcement learning
- active exploration
- function approximation
- long term
- action selection
- exploration strategy
- learning algorithm
- exploration exploitation
- active learning
- state space
- temporal difference
- model free
- real world
- model based reinforcement learning
- autonomous learning
- reinforcement learning algorithms
- optimal policy
- machine learning
- hamming distance
- hash functions
- temporal difference learning
- frame rate
- multi agent
- robotic control
- data sets